CT辅助消化道胶囊内窥镜图像虚拟外翻

CT辅助消化道胶囊内窥镜图像方法的主要目的在于更为直观地提高胶囊内窥镜的小病灶检出率,减少胶囊内窥镜的重复检查次数,减少患者检查费用,提高医生的工作效率.现有的虚拟外翻技术主要有两种,一种是展平法,一种是镜像法.CT辅助消化道胶囊内窥镜虚拟外翻是以镜像法为指导思想对三维肠道模型进行外翻,实验结果证明了这一方法的有效性.三维图像的显示采用了经典的移动立方体的算法.验证完成后,算法通过TI DaVinciDSP异构双核处理器(DM6467)和接口进行加速实现.通过合理分配双核的任务,算法的运算效率得到了很大提高.     

一种细胞核内DNA物质含量准确测量及回归校正新方法

在定量细胞学研究中,细胞核内DNA物质含量的准确测量是癌症筛查与病理诊断的必要前题与最重要依据。由于算法、设备、环境等因素的影响,在对细胞核的数字显微图像进行处理与分析、测量DNA物质含量时会产生较大的误差。本文提出了一种基于数学形态学和k近邻回归算法的DNA物质含量校正新方法。该方法首先利用膨涨算法对细胞核分割掩码进行处理,从而对DNA物质含量的测量进行空间校正;然后采用k近邻回归算法,充分利用细胞核的形态、纹理等特征参数所蕴含的信息,从而对DNA物质含量进行光学回归校正。实验表明,该方法能够显著提高DNA物质含量测量的准确性和可信度,对提高病理诊断的特异性与敏感性都有积极的意义。     

Effect of slow display on detectability when browsing large image datasets

Abstract— In this work, the effect of display temporal characteristics on detectability of signals was studied by comparing detection performance with a slow liquid‐crystal‐display (LCD) monitor and a fast, cathode‐ray‐tube (CRT) monitor when browsing multi‐slice image datasets in stack‐mode presentation. Thirteen readers evaluated 200 image sequence pairs in a two‐alternative forced choice experiment. The image sets consisted of three‐dimensional cluster lumpy backgrounds and were presented to the readers in two display devices: a three‐million‐pixel medical color LCD and a color desktop CRT. For the LCD, many inter‐gray‐level transitions are on the order of 50–60 msec, which was almost twice the frame time. The CRT had 2–5‐msec inter‐gray‐level transitions. The reader study was performed with a graphical‐user interface programmed using direct calls to OpenGL libraries to precisely control the browsing speed. The results were analyzed in terms of the difference in reader performance for each reader and each display, between a browsing speed of 20 and 50 frames per sec (fps). Average reader performance difference between 20 and 50 fps was measured to be 0.049 and 0.156 for the CRT and LCD monitors, respectively. The corresponding drop in reader performance associated with slow display was 0.11.     

一种用于半透明体绘制的光学模型

为了高效率地重建3维医学图像,并能充分显示内部隐含分界面及内部细节的详细信息,提出了一种能在普通硬件条件下实现的用于半透明体绘制的光学模型IVROM(improved volume rendering optical model)。该模型在光线吸收与发射模型的基础上,考虑了阴影、直接散射与间接散射等因素。此外,还详细介绍了结合该模型与Shear-Warp算法的半透明体绘制方法的实现。实验结果表明,采用该模型的半透明体绘制方法不仅效率高,而且显示效果好。     

Introduction to the special section on contemporary issues in human sexuality: research and practice

Critical issues in human sexuality have changed markedly during the past decade and so have the contributions of psychology to the evaluation and treatment of sexual behavior problems. Some of the greatest challenges facing psychologists today include the increasing emphasis on medical interventions for the male sexual disorders and the relative decline of traditional behavioral sex therapies, the recalcitrance of sexual desire disorders, the prevention of "unsafe" sexual behavior, and the limited understanding of female sexuality. This overview introduces a series of 5 articles that examine these selected critical issues in human sexuality. Guidelines for clinical practice and directions for future research are highlighted.     

基于改进3D UNet的前列腺MR图像分割

针对前列腺磁共振 (magnetic resonance, MR)图像边缘模糊、对比度较低,灰度值分布不均衡而导致分割精度较差的问题,提出了一种结合双路径注意力(dual path attention,DPA) 和多尺度特征聚合(multi-scale feature aggregation,MFA) 模块的改进3D UNet网络模型。首先,对数据集进行重采样和裁剪处理以适应模型输入。然后,在3D UNet网络的编码器各层引入DPA 并添加残差连接,加强特征的 编码能力。同时,在网络解码器中加入MFA模块,以充分利用空间上下文信息,增强语义信息。最后,在公开数据集PROMISE12上进行验证,所提出的模型的Dice系数为89.90%,Hausdorff 距离为9.37 mm。相比较于其他模型,所提出模型的分割结果更优,且参数量和运算量更少。     

Horizontal Voting Ensemble Based Predictive Modeling System for ColonCancer

Colon cancer is the third most commonly diagnosed cancer in the world. Most colon AdenoCArcinoma (ACA) arises from pre-existing benign polyps in the mucosa of the bowel. Thus, detecting benign at the earliest helps reduce the mortality rate. In this work, a Predictive Modeling System (PMS) is developed for the classification of colon cancer using the Horizontal Voting Ensemble (HVE) method. Identifying different patterns in microscopic images is essential to an effective classification system. A twelve-layer deep learning architecture has been developed to extract these patterns. The developed HVE algorithm can increase the system’s performance according to the combined models from the last epochs of the proposed architecture. Ten thousand (10000) microscopic images are taken to test the classification performance of the proposed PMS with the HVE method. The microscopic images obtained from the colon tissues are classified into ACA or benign by the proposed PMS. Results prove that the proposed PMS has ~8% performance improvement over the architecture without using the HVE method. The proposed PMS for colon cancer reduces the misclassification rate and attains 99.2% of sensitivity and 99.4% of specificity. The overall accuracy of the proposed PMS is 99.3%, and without using the HVE method, it is only 91.3%.     

基于STM32和小波自适应滤波算法的生理参数监测系统的研究

针对心血管疾病的高死亡率以及人口老龄化的现象,本篇文章开发了基于STM32单片机和小波自适应阈值滤波算法的可穿戴式健康监测系统。系统可分为系统微处理器、数字系统模块、人机交互模块、信号采集模块和无线通信模块等几个部分,针对人体的心率、血氧、体温等重要生理参数进行处理分析,进而对人体实时监护。系统处理器选取STM32F103C8T6作为控制芯片,显示模块选用了OLED。生理参数采集系统选用了MAX30102传感器和Pulse sense传感器分别对人体腕部和指尖心率进行采集。生理参数采集完毕后,通过进一步的A/D转化,基于提出的一种改进小波自适应阈值滤波算法降噪滤波,从而将人体的生理特征参数记录下来。再将采集的生理数据通过蓝牙传输至手机端,其中的ZigBee模块主要是把获得的数据再次输送到远程控制端内,让患者能够远程得到更好的医疗监控。本系统通过软件与硬件相结合的方式。最后通过对比论证其中心率（BPM）结果误差为±2BPM,血氧含量监测结果误差在±2%以内。     

基于深度学习的病历质量控制系统设计

医疗领域患者的主诉信息是医疗文本分类工作的关键,能为智慧医疗和信息文本归类提供有力的支持。近几年来随着深度学习的发展应用,基于传统深度学习技术的全流程病历内涵质量控制模型层出不穷,但传统模型存在很多缺点和局限性,诸如训练速度慢、精度损失、过拟合和无法处理大规模数据的问题,因此,引入改进的深度学习算法。指南指导下基于深度学习的全流程病历内涵质量控制体系实验结果为,将词向量设置成160时双向循环神经网络(Bidirectional Recurrent Neural Network,BiGRU-SA)模型效果最优,准确率为84.9% 。BiGRU-SA MODEL,精准度受向量维度的影响并不大。而改进的文本分类式前馈神经网络(Transformation-extraction-convolutional CNN,TextCNN)模型,精准度在其进行第3次和第四次迭代更新时,发生指数级增长,并在第3次迭代时,精度达到理想值,为8.3×10_-1^。随着迭代次数的增加,模型准确率呈现先增大后减小的趋势,在进行第6次迭代时模型效果最优,准确率为84.9% 。优化后的全流程病历内涵质量控制模型在变动率指标下的面积的值、准确率、F1、召回率四项指标值都有了一定的提升,以上结果能更好地解决过拟合和特征信息丢失的问题,并且实现全流程病历内涵质量的控制。     

无刷电机驱动器在大型医疗设备上的应用

无刷电机及其驱动器在大型医疗上的设备应用越来越广泛。为了设计更好的医疗设备、造福患者,研发人员需要对驱动器有更深的理解和更好的选型及应用。需要理解无刷电机及驱动器的工作原理,清楚大型医疗设备对无刷电机驱动器的要求和实现方式,如安全性、可靠性、通讯等。在医疗设备开发中,驱动器的应用有多种方案,它们各具特点,包括通用驱动器,板载驱动器、自主研发驱动器。随着技术的发展,无刷电机驱动器在大型医疗设备的应用,出现了新的趋势,如:小型化、一体化、模块化和集成化。